Cell biologists have a lot to thank jellyfish for. The species concerned has provided them with green fluorescent protein (GFP), a tool for monitoring the dynamics of gene expression and following the movement of proteins and even entire cells. Over the years, the usefulness of this 'molecular reporter' has been boosted by the isolation of GFP variants possessing greater fluorescence, altered excitation and emission wavelengths, and greater stability in living cells.

But there was scope for further improvement. In particular, tracking protein movement in real time was difficult if the tagged protein became evenly distributed in the cell under steady-state conditions. Writing in Science (297, 1873–1877; 2002), George H. Patterson and Jennifer Lippincott-Schwartz now describe a possible solution to this problem. They have produced a GFP variant that shows greatly increased fluorescence if it is activated by light; and it functions in physiological conditions. Using this photoactivatable GFP, tagged proteins in a small area of the cell can be selectively marked by light activation, so that their movement through the cell can be followed against a dark background by fluorescence microscopy.

Normal GFP contains a mixed population of neutral and anionic chromophores — molecules that absorb light then re-emit it. These are respectively associated with a major light-absorption peak at a wavelength of 397 nm and a minor peak at 475 nm. Intense illumination at 400 nm shifts the population to the anionic form, thereby increasing the absorbance of the minor peak. Patterson and Lippincott-Schwartz set out to develop a GFP variant that had a negligible 475-nm peak. They reasoned that illumination at 400 nm would then produce a much greater proportional increase in 475-nm absorbance compared with the normal protein, and therefore increase optical contrast.

One variant, called PA (for 'photoactivatable') GFP, produced a 100-fold increase in fluorescence at 488 nm in vitro, and was stable under various conditions. Patterson and Lippincott-Schwartz then tested this variant in living cells. They found that selective photoactivation of PA-GFP in either the nucleus or cytoplasm of the cell led to a brightly fluorescent population that moved rapidly between the two areas. The authors also attached PA-GFP to a marker protein called lgp 120, found in a cellular compartment, the lysosome, that is responsible for digesting unwanted material. Before photoactivation, little fluorescence of lgp 120 was seen. But after photoactivation, lgp 120 was evident in most of the lysosomes within 20 minutes (see figure), showing that rapid exchange occurs between these organelles.

This new type of GFP should prove valuable for studying the temporal and spatial dynamics of proteins in the living cell. The future for such research literally looks brighter.